Abstract

This brief argues that charter school programs can have direct fiscal impacts on school districts for two reasons. First, operating two systems of public schools under separate governance arrangements can create excess costs. Second, charter school financing policies can distribute resources to or away from districts. Using the city school districts of Albany and Buffalo in New York, we demonstrate how fiscal impacts on local school districts can be estimated. We find that charter schools have had fiscal impacts on these two school districts. Finally, we argue that charter schools policies should seek to minimize any avoidable excess costs created by charter schools and ensure that the burden of any unavoidable excess costs is equitably distributed across traditional public schools, charter schools, and the state. We offer concrete policy recommendations that may help to achieve these objectives.

Introduction

Charter schools have been a rapidly growing part of U.S. education for two decades, and the U.S. Department of Education's Race to the Top initiative is likely to spur continued growth. Since the birth of charter schools, concerns have been raised that they would drain resources from traditional public schools (Molnar 1996; Arsen, Plank, and Sykes 1999). Given the budgetary strain that school districts have been facing in recent years and the impetus to increase the number of charter schools, concerns about the fiscal impacts of charter schools are more salient than ever. Charter school policy debates often mention potential fiscal impacts on local public schools, but relatively little research has focused on this issue.1

This brief addresses this gap in the literature by demonstrating how fiscal impacts can be estimated and offers a way to conceptualize fiscal impacts that is useful for framing charter school policy objectives. This brief will not assess the overall merits of charter schools. For instance, fiscal pressure created by charter schools might spur a school district to improve their programming, and such competitive pressure has the potential to more than offset any negative effects of reduced resources. Our analysis does not try to estimate such effects of charter schools on educational quality.2 Rather, we explore how states’ charter school policies can affect district revenues and expenditure needs. Our policy recommendations focus on how to: (1) reduce avoidable excess costs created by charter schools, and (2) ensure that the burden of any unavoidable excess costs is equitably distributed across traditional public schools, charter schools, and the state. If a state chooses to allow charter schools, then traditional public schools and charter schools should be given incentives to engage in efficient collaborative activities, as well as opportunities to compete for students on a level playing field. We offer concrete policy recommendations that may help to achieve these objectives.

Charter school programs can have fiscal impacts on districts for indirect reasons, often referred to as systemic or general equilibrium effects. By breaking the link between residential location and school services, school choice programs can influence household location decisions and thereby alter property values.3 Because many districts rely heavily on local property taxes, changes in property values can alter districts’ capacities to generate revenue. In addition, charter schools might influence the political economy of education finance within a school district. If fewer families residing in a district send their children to district schools, then political support for taxes to finance those schools may diminish. Alternatively, in states where local tax dollars are passed through to charter schools, charter schools might increase political support for taxes because they draw students away from private schools. The size and direction of indirect or systemic effects of charter schools on school district finances are thus ambiguous.

In this brief, we focus on two direct effects of charter schools on school district budgets. First, operating two systems of public schools under separate governance arrangements can create excess costs, some of which are incurred by the district. Second, charter school financing policies can distribute resources to or away from districts. We begin by defining the concept of excess costs and identifying the types of costs incurred when charter schools enter a school district. Next, we identify the ways that charter school finance policies can redistribute revenues either to or away from local schools districts. Third, we summarize our attempts to estimate the fiscal impacts of charter schools on two local school districts in New York—the city school districts of Albany and Buffalo. Albany and Buffalo are interesting case studies because they have relatively large concentrations of charter school students and they have, respectively, stagnant and shrinking enrollment bases. Thus, we would expect larger fiscal impacts of charter schools in these districts than most other places, providing an indication of how large charter school fiscal impacts might be. Finally, we discuss how charter school financing policies can help to address fiscal impacts by creating incentives to minimize avoidable excess costs and by fairly distributing the burden of the unavoidable excess costs that remain.

Excess Costs Generated by Charter Schools

Charter schools can increase educational costs either by causing an increase in revenues devoted to education or by causing a reduction in services. If the revenues devoted to education are increased, an additional burden is placed on taxpayers. Reductions in services represent a burden borne primarily by students and their families. We refer to the burdens created for taxpayers and students as a result of charter schools as “excess costs.” This term does not imply the overall costs from charter policies outweigh the benefits. There may very well be benefits that exceed the excess costs. Furthermore, additional costs in some areas may be offset by cost savings in other areas, especially if schools allocate resources more efficiently.4 Regardless of these potential benefits created by charter schools, it is important to understand the excess costs created by charter schools.

Charter schools can generate excess costs for a number of reasons. First, charter schools can be expected to attract some number of students from private schools (Toma, Zimmer, and Jones 2006; Ludner 2007; Chakrabarti and Roy 2010; Buddin 2012). The additional resources that charter schools use to educate these students are not necessarily new resources from the point of view of society. Nonetheless, transfers from private to charter schools do shift educational costs from the private schools and their parents to the public sector and taxpayers, thus creating fiscal impacts for public education systems.

Second, charter schools might cause more personnel resources to be used to educate a given number of students. In our case studies of Buffalo and Albany, district officials indicated that it is difficult to reduce the number of teachers when enrollment losses are spread across a large number of schools and grades. For instance, if five students are the most any particular grade in a school loses to charter schools, it might not be possible to reduce the number of classroom teachers in the district.5 In this case, the additional teachers hired by charter schools would not be offset by reductions in the number of district teachers. More generally, uncertainty about charter school enrollments can make it difficult for districts to project enrollments and to maintain targeted class sizes. If districts are apprehensive about exceeding class size targets, then they will tend to err on the side of smaller classes. Although class size reductions may be beneficial to students, the benefits generated might not be sufficient to offset the costs to taxpayers or the reductions in other areas of the school budget required to finance the reduced class sizes.

Third, charter school entry into a district typically increases the number of school buildings used to serve students, increasing facility and related maintenance costs. Closing a school in a district is politically contentious and is typically not feasible until enrollment losses are sufficiently large in particular grade ranges and expected to persist. Also, districts may need to maintain excess physical capacity in case charter schools close and enrollments suddenly increase. In 1999, the year when the first charter school opened in Albany, the Albany City School District served 10,380 students in 17 schools. As of 2009–10, public resources supported 24 schools (15 district and 9 charter schools) that served 10,568 students.

Finally, in many states, districts are required to provide several services for charter school students including transportation, special education evaluation, and health services. The costs of these services might be higher as a result of charter schools because the services have to be delivered to students spread across a larger number of schools and locations. For instance, the district may need to establish additional bus routes to transport students to charter schools. In addition, administering charter school payments, coordinating special education services, and addressing other coordination issues with charter schools increases demands on district offices. For example, the Chief Financial Officer for Buffalo Public Schools indicated that one of her staff members works full time on charter school payment issues.

These are all excess costs created by the introduction of charter schools, and many of them result from coordination difficulties associated with separately operating two systems of publicly funded schools. It might be possible to reduce some of these costs through improved planning by the district or improved coordination between the district and charter schools. In principle, then, one may distinguish avoidable excess cost from unavoidable excess costs.6

Impacts of Charter School Finance Policies

The amount of excess costs generated by charter schools needs to be weighed against any benefits of charter schools in comprehensive policy assessments. The fiscal impacts of charter schools on local school districts, however, depend not only on the amount of excess costs but also on the redistribution of revenues that results from charter school financing policies, which vary considerably across states (Goldhaber et al. 2005; Ni and Arsen 2010).

To understand how charter school policies redistribute revenues, it is useful to distinguish two basic approaches to charter school financing. One approach is for the state to make per pupil payments directly to the charter school without any involvement of local school districts. Under this approach, district residents who enroll in charter schools would typically not be included in enrollment counts for purposes of determining most state aid awards. Thus, charter school payments are at least partially financed by reductions in state aid to districts, and the primary effect of charter schools on school districts is reduced state aid receipts. The second approach is for local school districts to make payments to charter schools for each resident student that enrolls in a charter school. Under this approach the charter school students who reside in the district would typically continue to be included in district enrollment counts, and thus, charter school enrollments would not reduce state aid awards to the district. The primary fiscal impacts on charter schools are the payments they are required to make to charter schools.

When charter school payments are made directly by the state, fiscal impacts will depend on how much a district relies on state aid. In districts that receive only small amounts of aid, charters will be likely to shed costs at least as great as revenue losses resulting from charter school enrollments. In this case, the burden of any excess costs generated by charter schools will be split between the state and charter schools, and the split will depend on the size of charter school payments. In districts that rely heavily on state aid payments, the fiscal impacts of charter schools will be greater, and the students and taxpayers in these districts may bear some of the burden of the excess costs generated by charter schools.

In places where charter school payments are made by districts, fiscal impacts will naturally depend on the size of the payments made to charter schools. If per pupil charter school payments are low relative to per pupil spending in a district, fiscal impacts will be minimized and much of the cost of additional resource usage caused by charter schools will be borne by charter school students in the form of reduced services. Alternatively, high charter school payments will increase fiscal impacts and force much of the excess cost created by charter schools onto local public school students and taxpayers.

The effect of transfers from private schools into charter schools will also depend on the school finance regime. Where the state makes charter school payments directly, transfers from private schools will have no direct effect on district budgets, although they will place increased burdens on the state. Where districts make charter school payments, the effects are more complicated. The enrollment of private school students will increase the payments that districts need to make to charter schools, but will also increase the count of public school students who reside in the district and thereby increase state aid awards to the district. The net effect on the district will depend on the relative sizes of its per pupil aid awards and charter school payments.

Finally, state financing programs can serve to disperse the costs created by charter schools to taxpayers statewide. For instance, New York State provides districts with increasing charter school enrollments transitional aid meant to reduce fiscal impacts on the district. The state also provides limited start-up and facility grants to charter schools, essentially passing some of the costs of excess facility capacity onto taxpayers statewide.

Estimating Fiscal Impacts in Buffalo and Albany

Buffalo and Albany are interesting case studies because they have a high concentration of charter schools. Approximately 20 percent of public school students in Albany and 17 percent in Buffalo attend charter schools. These market shares, respectively, rank tenth and fifteenth highest among districts nationwide (NAPCS 2011). Also, the school-aged population in Albany has been stagnant for the last decade and the school-aged population in Buffalo has been falling for at least two decades. Transfers to charter schools thus result in shrinking enrollments in these districts rather than merely slower enrollment growth.7 Finally, state law in New York requires districts to pay charter schools an amount equal to per pupil operating expenditures for each resident student who enrolls in a charter school. Due to each of these factors, fiscal impacts of charter schools are likely to be larger in Buffalo and Albany than in most other locations.

A straightforward way to estimate the fiscal impacts of charter schools is to compare the change in district revenues net of charter school payments to the expenditure reductions that the districts are able to make as a result of charter school enrollments. The changes in both revenues and expenditures depend on the impacts that charter schools have on district enrollments. So we begin by presenting estimates of enrollment impacts. Next, we estimate the changes in revenue and expenditures generated by these enrollment decreases.

Enrollment Impacts

Data on district enrollments were drawn from the New York State Education Department's (NYSED) School Report Cards, and counts of district residents attending charter schools were provided by the NYSED Office of State Aid. As shown in figure 1, between 1999 and 2009, K–12 enrollment decreased 19.4 percent in Albany and 26.7 percent in Buffalo. The enrollment of students in district schools plus the charter schools located in Albany has remained roughly flat over the last 10 years. Nonetheless, the district saw significant declines in enrollments, especially between 2002 and 2008 as charter schools in the district were expanding. In Buffalo, enrollment declines had already begun during the 1990s, prior to charter schools, but have accelerated over the last decade, particularly between 2002 and 2005.

Figure 1.

K–12 Enrollment in Albany and Buffalo, 1990-2009. Source: National Center for Education Statistics Common Core of Data and New York State School Report Cards, Accountability and Overview Reports.

Figure 1.

K–12 Enrollment in Albany and Buffalo, 1990-2009. Source: National Center for Education Statistics Common Core of Data and New York State School Report Cards, Accountability and Overview Reports.

During 2009–10, 2,054 students residing in Albany attended charter schools, approximately 20 percent of public school enrollment. In Buffalo, 6,557 resident students attended charter schools in 2009–10, approximately 17 percent of public school enrollment. Relatively small percentages of limited English proficient students from these districts are enrolled in charter schools, and charter schools in Albany enroll few students with disabilities (see first column, Scenario 1, in table 1), which limits the ability of districts to reduce spending on bilingual, English as Second Language, and special education services. Charter schools in Albany, however, serve a somewhat higher percentage of low-income students than the district, which might help to reduce compensatory education needs in the district.

Table 1.
Estimated Impact of Charter Schools on District Enrollment, 2009–10
Scenario 1Scenario 2
#% of Public Enrollment#% of Public Enrollment
 Albany 
Totala 2,054 20.5 1,837 18.7 
Free lunch eligibleb 1,655 29.2 1,524 27.5 
Limited English proficientb 50  9.1 39  7.2 
Students with disabilitiesb 89  6.1 80  5.5 
 Buffalo 
Totala 6,557 16.7 5,832 15.2 
Free lunch eligibleb 4,011 14.9 3,829 14.3 
Limited English proficientb 107  3.4 98  3.1 
Students with disabilitiesb 855 12.2 808 11.6 
Scenario 1Scenario 2
#% of Public Enrollment#% of Public Enrollment
 Albany 
Totala 2,054 20.5 1,837 18.7 
Free lunch eligibleb 1,655 29.2 1,524 27.5 
Limited English proficientb 50  9.1 39  7.2 
Students with disabilitiesb 89  6.1 80  5.5 
 Buffalo 
Totala 6,557 16.7 5,832 15.2 
Free lunch eligibleb 4,011 14.9 3,829 14.3 
Limited English proficientb 107  3.4 98  3.1 
Students with disabilitiesb 855 12.2 808 11.6 

Notes:aScenario 1 represents total enrollment of resident students in charter schools provided by NYSED Office of State Aid. Scenario 2 represents estimates that assume that in the absence of charter schools, charter school students would have enrolled in private schools at the same rates as resident students with a similar poverty, English language proficiency, and disability profile. Rates of private school attendance for each district (by poverty status, English language background, and disability status) were obtained from the 2000 U.S. Census.

bFigures in these rows are estimates computed by obtaining the percentage of students enrolled in each category in charter schools located in the district from the School Report Cards and then multiplying that percentage by the total count of resident students enrolled in charters.

Resident charter school enrollments are upper-bound estimates of the numbers of students districts have lost to charter schools. Some of the students who reside in a district and attend charter schools might have chosen to attend private schools or to move out of the district if charter school options were not available. To address the uncertainty regarding how many charter school students would have attended district schools in the absence of charters, we developed estimates of fiscal impacts under two scenarios. Under Scenario 1, we assume that all charter school students would have attended district schools if charter school options were not available. Under Scenario 2, we assume that in the absence of charter schools, charter school students would have enrolled in private schools at the same rates as resident students with a similar poverty, English language proficiency, and disability profile. Rates of private school attendance for each district (by poverty status, English language background, and disability status) were obtained from the 2000 U.S. Census. Prior research suggests the reality in Albany and Buffalo should be somewhere between these two scenarios. For example, Chakrabarti and Roy (2010) find that Michigan charter schools reduced local private schools’ enrollments but at lower rates than they reduced public schools’ enrollments. The estimated impacts on district enrollments under the two scenarios are presented in table 1.

Fiscal Impacts

Charter school enrollments affect net district revenues in New York State in three ways. First, districts are required to make payments to charter schools for each resident charter school student. Second, awards from Title I, Part A, the largest single federal aid program, are determined based on district enrollments, excluding charter school students. Thus, when Title I students who otherwise would attend district schools choose to enroll in a charter school, the federal aid received by the district is reduced.8 Third, other state and federal aid programs base awards on counts of resident pupils in the district, including charter school students. Charter school students who are drawn from private schools or who would otherwise have moved out of the district generate additional aid the district would not have received in the absence of charter schools.

Data on revenues and charter school payments used in our analysis were drawn from Annual Financial Reports (ST-3 files) collected by the NYSED. The second rows of each panel of table 2 display our estimates of revenue losses. We estimate that the Albany City School District lost between $24.9 and $26.1 million (between 11.9 and 12.5 percent of total revenues) during 2009–10 as a result of charter schools, and the Buffalo Public Schools lost between $67.0 and $76.8 million (between 8.6 and 9.9 percent of total revenues) in the same time period. The range of estimates is wider for Buffalo because it relies more heavily on state aid than Albany, and therefore variation in assumptions about how many charter school students would have attended district schools has a larger effect on state aid awards.

Table 2.
Net Fiscal Impacts of Charter School Enrollments, 2009–10
Scenario 1Scenario 2
 Albany 
Lost enrollment 2,054 1,837 
Lost revenue, $ (26,138,249) (24,864,664) 
Reduced expenditure need, $ 19,723,874 17,643,668 
Net fiscal impact, $ (6,414,955) (7,220,996) 
Net fiscal impact per pupil, $ (804) (905) 
Transition aid, $ 6,179,250 6,179,250 
Transition aid per pupil, $ 711 711 
 Buffalo 
Lost enrollment 6,557 5,832 
Lost revenue, $ (76,801,485) (67,036,444) 
Reduced expenditure need, $ 52,812,499 43,467,775 
Net fiscal impact, $ (23,988,986) (23,568,669) 
Net fiscal impact per pupil, $ (736) (723) 
Transition aid, $ 4,634,647 4,634,647 
Transition aid per pupil, $ 132 132 
Scenario 1Scenario 2
 Albany 
Lost enrollment 2,054 1,837 
Lost revenue, $ (26,138,249) (24,864,664) 
Reduced expenditure need, $ 19,723,874 17,643,668 
Net fiscal impact, $ (6,414,955) (7,220,996) 
Net fiscal impact per pupil, $ (804) (905) 
Transition aid, $ 6,179,250 6,179,250 
Transition aid per pupil, $ 711 711 
 Buffalo 
Lost enrollment 6,557 5,832 
Lost revenue, $ (76,801,485) (67,036,444) 
Reduced expenditure need, $ 52,812,499 43,467,775 
Net fiscal impact, $ (23,988,986) (23,568,669) 
Net fiscal impact per pupil, $ (736) (723) 
Transition aid, $ 4,634,647 4,634,647 
Transition aid per pupil, $ 132 132 

The key question for determining fiscal impacts is whether enrollment reductions allow a district to achieve expenditure reductions commensurate with revenue reductions. To address this question we used detailed reports of expenditures by function to divide district spending into fixed components that cannot be adjusted in response to enrollment losses, and step or variable costs, which can be adjusted. Next, for each variable or step cost we identified key cost drivers, either the number of students or number of teachers, which allowed us to compute per unit cost figures. Finally, we used estimates of enrollment changes, under Scenario 1 and Scenario 2, together with per unit cost figures to estimate how much expenditures could be reduced. Data on expenditures by function were drawn from the ST-3 files and counts of teachers were drawn from the Personnel Master File maintained by the NYSED.

For these purposes, fixed costs include spending for services that districts have to provide for resident students, regardless of whether they attend district or charter schools. In New York, these services include textbooks, computer hardware and software, pupil transportation, pre-K services, and health services. Also included among fixed expenditures is spending that is difficult to reduce when enrollments decrease, at least in the short run. These include expenditures for general support functions, principals, capital expenditures, debt service, special schools, community services, and retiree health care benefits. The remaining expenditure items we count as variable costs, including spending on instructional supervision other than principals, teachers, pupil services other than health, employee preparation programs, and occupational education. For Albany, 66.3 percent of the expenditures are counted here as variable, and for Buffalo, 54.6 percent. The appendix provides details on our categorization of various types of expenditures and the cost drivers used to calculate the cost savings permitted by enrollment losses.

The third rows of each panel of table 2 report our estimates of cost savings. We estimate that charter school enrollments allowed the Albany City School District to reduce expenditures by between $17.6 and $19.7 million, which is between 9.1 and 10.0 percent of the expenditures that would have been required in the absence of charter schools. For the Buffalo Public Schools, we estimate that charter school enrollments allowed expenditure reductions between $43.5 and $52.8 million, which is between 6.0 and 7.2 percent of total expenditures.

The reduced expenditure needs reported in table 2 are based on the average variable costs of students drawn away from district schools to charter schools. Variable costs vary with the number of students served and represent the costs that can be avoided when students transfer to a charter school. The variable cost associated with each student depends on that student's educational needs. For instance, limited English proficient students and students with disabilities require more resources than the typical student. Also, student teacher ratios and spending on programs such as vocational education and athletics are different at different grade levels. Table 3 presents our estimates of the variable costs associated with six different types of students defined by grade level, limited English proficiency, and disability status.9 Our estimated reductions in expenditure needs in table 2 are based on average variable costs given these estimates in table 3 and the actual mix of charter school students in Albany and Buffalo. If additional charter schools were to attract greater proportions of limited English proficient (LEP), disabled, or high school students than do the current charter schools, then the per pupil reductions in the districts’ expenditure needs would be larger.

Table 3.
Variable Costs (in $) Associated with Different Student Classifications
KindergartenGrades 1–6Grades 7–12
 Albany 
Regular education  7,649a  9,338b  9,856c 
Limited English proficient  9,604d 11,293d 11,810d 
Students with disabilities 19,583e 21,272e 21,790e 
 Buffalo 
Regular education  5,213a  5,893b  8,194c 
Limited English proficient  7,715d  8,395d 10,695d 
Students with disabilities 16,961e 17,641e 19,941e 
KindergartenGrades 1–6Grades 7–12
 Albany 
Regular education  7,649a  9,338b  9,856c 
Limited English proficient  9,604d 11,293d 11,810d 
Students with disabilities 19,583e 21,272e 21,790e 
 Buffalo 
Regular education  5,213a  5,893b  8,194c 
Limited English proficient  7,715d  8,395d 10,695d 
Students with disabilities 16,961e 17,641e 19,941e 

Notes:aComputed by summing the variable cost categories marked “Teachers” or “All students” in the Cost Driver column in the Appendix, dividing by total district enrollment, and then adding the variable costs marked by “K-students” divided by the district's kindergarten enrollment. Personnel costs categories are multiplied by a fringe benefit factor before being divided by student counts.

bComputed by summing the variable cost categories marked “Teachers” or “All students” in the Cost Driver column in the Appendix, dividing by total district enrollment, and then adding the variable costs marked by “G1-G6 stu” divided by the district's grade 1 through grade 6 enrollment. Personnel costs categories are multiplied by a fringe benefit factor before being divided by student counts.

cComputed by summing the variable cost categories marked “Teachers” or “All students” in the Cost Driver column in the Appendix, dividing by total district enrollment, and then adding the variable costs marked by “G7-G12 stu” divided by the district's grade 7 through grade 12 enrollment. Personnel costs categories are multiplied by a fringe benefit factor before being divided by student counts.

dComputed by adding the variable costs marked by “LEP students” in the Cost Driver column multiplied by a fringe benefit factor and divided by the count of LEP students in the district to the variable cost per regular education student for the same grade level.

eComputed by adding the variable costs marked by “Stu. w. Disabil.” in the Cost Driver column multiplied by a fringe benefit factor and divided by the count of special education students in the district to the variable cost per regular education student for the same grade level.

The estimated revenues lost due to charter schools by the Albany City School District exceed the estimated reduction in expenditure needs by between $6.4 and $7.2 million, depending on how many charter school students would have otherwise attended district schools. These imply negative impacts between $804 and $905 per pupil enrolled in district schools. For the Buffalo Public Schools we estimate negative fiscal impacts between $23.6 and $24.0 million, or between $723 and $736 per pupil. In Albany, negative fiscal impacts increase as the number of charter school students drawn from the private sector increases because the additional state aid those students generate does not offset the additional payments to charter schools. In Buffalo, however, per pupil state aid amounts are higher than in Albany, and additional students drawn from private schools reduce estimated fiscal impacts in Buffalo.

The sixth row of each panel of table 2 lists the transition aid that New York State provided to the districts to reduce the fiscal burden of their charter school payments. We discuss this transition aid in more detail in the next section. In Albany, this aid amounted to $711 dollars per pupil and offset between 78 and 88 percent of the estimated negative impact. For Buffalo, transition aid amounted to $132 per pupil and offset about 18 percent of the estimated impact.

The negative fiscal impacts in Albany and Buffalo suggest these districts have either had to find ways to reduce costs that our analysis treats as fixed or to decrease service levels as a result of charter school enrollments. As charter school enrollments level off and transition aid is correspondingly reduced, these cuts in fixed costs and service levels will need to become larger.

Additional Considerations

The primary way that a district can reduce costs that our analysis treats as fixed is by closing schools. Closing a school permits immediate reductions in expenditures on energy, maintenance, and janitorial staff. In addition, closing a school allows a district to reduce the number of principals and clerical staff, and facilitates achieving the teaching and staff reductions that our analysis already assumes can be achieved in response to enrollment losses. In the case of Albany, which is an independent school district which owns its school buildings, the district can also benefit financially by leasing or selling a closed facility. In the case of a fiscally dependent school district like Buffalo, where facilities are owned by the city rather than the district, it is not clear whether the district would benefit from leasing or selling closed schools.

Albany recently closed one middle school, and examination of excess capacities indicates that it may have room to close two elementary schools as a result of decreased enrollments. In 2009–10, Albany spent on average $821,263 per school for plant operations and maintenance.10 In addition, average salary and benefits for a principal in 2009–10 was $145,932, which suggests closing a school could save the district between $900,000 and $1 million per year.11 If these cost-saving estimates are correct, and we attribute all of the enrollment reductions the Albany school district has experienced over the last decade to charter schools, then charter school enrollments would allow the district to reduce expenditures by between $2.7 and $3.0 million in addition to the expenditure reductions summarized in table 2. Put another way, closing schools would allow the district to offset between 37 and 47 percent of the fiscal impact estimated in table 2.

In Buffalo, the district has closed a number of schools over the last decade. We estimate that charter school enrollments have allowed the district to close seven or eight more schools than it otherwise could have. In 2009–10, the Buffalo City School District spent on average $1,063,517 per school for plant operations and maintenance.12 In addition, average annual salary and benefits for a principal in the district is $128,346. This suggests that a school closure could reduce the costs we have treated as fixed in our analysis by between $1 and $1.2 million. Thus, closing seven or eight schools may have saved the district between $7.0 and $9.6 million, reducing the negative fiscal impacts that we estimated for Buffalo by between 29 and 41 percent.

Although school closures can help districts reduce fiscal impacts, it is also important to note that our estimates of fiscal impacts are not estimates of how much the Albany and Buffalo districts have been able to reduce expenditures as a result of charter school enrollments. Rather, we have estimated expenditure reductions that may be possible without sacrificing the levels of service that districts provide. Actual reductions do not necessarily match possible reductions.

For instance, our analysis assumes that districts maintain constant student/staff ratios as enrollment declines. Between the 2001–02 and 2008–09 school years, enrollment in the Albany City School District fell by almost 18 percent, but the district made few staffing changes. As a result, student/teacher ratios decreased from 12.4 in 2001–02 to 10.6 in 2008–09. Ratios of students to other types of staff also were substantially lower in 2008–09 than in 2001–02. Buffalo saw the largest drops in enrollment during 2002 to 2006, during which time enrollment declined by an average of more than 5 percent per year. During this period, the district was able to cut staff, and as a result maintained similar staffing ratios each year during this period.

The experiences in Albany and Buffalo indicate that over some period of time, staffing can be adjusted in response to enrollment declines. They also suggest that Albany was slower than Buffalo to adjust staffing in response to charter school enrollments. There are three possible reasons why Buffalo was able to respond more quickly than Albany. First, Buffalo underwent enrollment declines during the 1990s, so district officials there may have gained experience managing enrollment reductions sooner than officials in Albany. Second, enrollment declines were larger in Buffalo than in Albany, so Buffalo may have reached thresholds that made staff reductions easier and more urgent earlier. Third, Buffalo was closing schools beginning earlier in the decade, and closing schools facilitates staff reductions. Albany did not close any schools until 2008–09, after which it began to make staff reductions.

Policy Implications

Our case studies suggest that charter schools can have negative fiscal impacts on school districts. Even assuming districts are able to make staffing reductions commensurate with enrollment losses, we estimate fiscal impacts ranging from $804 to $905 per pupil in Albany and $723 to $736 per pupil in Buffalo. In the long run, districts may be able to reduce some of the expenditures that our analysis treats as fixed, and thereby mitigate some of the fiscal impact. Several factors, however, hinder immediate expenditure reductions in response to enrollment losses, therefore, the short run impacts of charters might be larger than our estimates indicate.

Revenues diverted from school districts to charter schools exceed the costs that charter school enrollments allow districts to shed, thereby causing districts to bear part of the burden of the excess costs charter schools generate. This analysis suggests two objectives for policies to address charter school fiscal impacts. The first objective is to reduce the avoidable excess costs that operating separate school systems create. The second is to divide the burden of these costs sensibly between the districts and their students and taxpayers, the charter schools and their students, and the state and its taxpayers. In the remainder of this brief we discuss policies to promote these objectives.

Policies to Promote Coordination between Districts and Charter Schools

Many of the excess costs created when charter schools enter an area arise because of difficulty coordinating service delivery across separate systems of schools, resulting in duplication of services and failure to realize potential economies. Several measures can help encourage stronger coordination.

  1. Constrain the timing of charter school enrollments to facilitate budget planning. States should consider requiring charter schools to have earlier, binding admissions deadlines. Students should not be allowed to enter a charter school during the year unless they declare that intention by the deadline or are replacing other students from their district that declared by the deadline but later transferred back to a district school. Charter schools would have to send rosters of each district's resident students with signed declarations shortly after the deadline. Such requirements would reduce uncertainty regarding district enrollments and thus facilitate timely adjustment of staffing levels.

  2. Create incentives for districts and charter schools to share facilities. Charter schools frequently close, which creates incentives for districts to maintain excess capacity as a hedge against possible enrollment increases. Districts might be better positioned to close schools if they could retain flexibility to reopen those schools should the need arise. One way to reduce excess capacity while retaining flexibility is to lease excess space to charter schools. States offering aid to charter schools to help them finance facilities could create incentives for facility sharing by providing more favorable aid to charters that lease space from public school districts. Favorable aid terms for charter schools that lease from district schools can be justified because the charter schools would be assuming some risk by leasing rather than purchasing facilities. The ability to reduce expenditures and/or receive lease payments may provide districts sufficient incentives to share excess space with charter schools. If not, then the state might require districts with excess capacity to offer space to charter schools.

  3. Encourage districts to use existing intra-district choice programs to facilitate staffing adjustments. Like many urban districts, both Albany and Buffalo offer intra-district school choice. Such choice programs provide an opportunity to ensure that schools achieve class-size targets. The number of teachers allocated to each school should be coordinated with enrollment decisions to ensure that class sizes do not deviate from established targets. Such a policy might require leaving classrooms vacant in some buildings and redeploying them for other purposes, and it may also mean fewer students get their first choice of school—but such a policy can help ensure the district is able to make staffing reductions commensurate with enrollment losses.

Policies to Promote Equity without Sacrificing Efficiency

Two considerations are relevant for dividing the burden of excess costs sensibly. One, incentives to reduce avoidable excess costs and improve programs need to be maintained. For instance, it may not be optimal to shift the burden of all negative fiscal impacts from the districts to the state, because districts would no longer face pressure to make long-run expenditure reductions or program improvements in response to reduced enrollments. Two, districts should not be required to bear the burden of excess costs which they cannot do anything to address. These burdens put undue pressure on services in these districts. Several measures can help achieve these objectives.

  1. Link charter school payments to estimates of costs that the district can reduce in response to enrollment losses. Currently in New York, several types of fixed expenditures affect the size of district payments to charter schools. For example, current retiree health care expenditures are not affected by charter school enrollments and are costs that charter schools typically avoid. In 2009–10, the Buffalo City School District spent approximately $1,658 per pupil and the Albany City School District spent approximately $1,064 per pupil for current retiree health benefits. Removing these expenditures from the computation of charter school payments would substantially relieve much of the fiscal impact of charter schools on districts. Because districts can do little to control these costs, and charter schools do not typically have similar costs, this adjustment would serve to distribute cost burdens more fairly across sectors. If payment amounts linked to the costs that districts can shed when students transfer to charter schools are insufficient to support charter school operations, then the state can consider providing additional aid to charter schools. Charter school policies are established by states to promote statewide education goals, suggesting that taxpayers statewide should share the burden of any excess costs such policies generate.

  2. Vary required district payments based on the educational needs of students transferring into charter schools. The analysis herein points out that the variable costs associated with a student depends on the student's educational needs and will vary by factors such as grade level, limited English proficiency, and disability status. To ensure that payments to charter schools are commensurate with expenditures that the district can reduce in response to enrollment losses, it is important to vary charter school payments with these characteristics and the educational needs of the students who are transferring into charter schools.

  3. Provide transitional aid to districts experiencing large growth in charter schools and facing stagnant population growth. New York's transitional aid program reimburses the districts for a portion of their charter school payments that are attributable to recent increases in charter school enrollment. The award amounts are computed as 80 percent of the payments attributed to increased charter school enrollment during the last year, 60 percent of payments attributed to increases in charter school enrollments two years earlier, and 20 percent of the payments attributed to increases in charter school enrollments three years earlier. Previously in this brief we discussed reasons why, soon after new charter schools open, districts might have difficulty achieving staff reductions commensurate with enrollment losses. This aid program is well designed to provide relief to districts who are experiencing rapid increases in charter school enrollments and who may need time to adjust. Because adjustments only need to be made by districts that experience enrollment declines rather than merely slower rates of growth, states could target this assistance toward districts that would have stable or shrinking enrollment bases in the absence of charter schools. Because this aid phases out once charter schools have remained open for a few years, it does not undermine incentives for districts to adjust their spending or improve programming in the longer run.

Summary

Our analysis suggests that charter schools can create fiscal impacts on school districts, particularly in districts with rapid growth in charter schools and declining or stagnant enrollment bases. If districts are able to reduce costs that are fixed in the short run by closing schools and taking other measures, they might be able to avoid long-run reductions in service quality. Achieving expenditures reductions commensurate with enrollment and revenue losses is difficult in the short run, however, for a number of reasons. The primary reason why charter schools impose negative fiscal impacts is that running two systems of schools in parallel creates excess costs. Policies should be designed to minimize any avoidable excess costs that are created by the introduction of charter schools, and to ensure that districts, their taxpayers, and the charter schools themselves do not bear an undue portion of the excess costs. As discussed in this brief, there are a number of steps state policy makers can take to promote these policy goals.

Notes

1. 

A search of the ERIC, EconLit, and RePEc databases found just one peer-reviewed study that provides measures of the fiscal impact of charter schools. Arsen and Ni (2012) find that higher levels of charter school enrollments in Michigan school districts are strongly associated with declining fund balances, and that revenues declined more rapidly than costs in districts losing students to charter schools. A few other reports prepared by research centers or advocacy organizations have made arguments about likely fiscal impacts but do not actually attempt to estimate those impacts (see, for instance, Anderson 2004 and Little et al. 2003).

2. 

Studies of the impact of charter schools on student performance in traditional public schools face difficult methodological challenges, and those that have been conducted provide mixed results. A few studies have found positive effects of charter schools on traditional public schools (Sass 2006; Booker et al. 2008), but others have found negligible or even negative effects (Bettinger 2005; Bifulco and Ladd 2006; Imberman 2011).

3. 

Theoretical simulations suggest that providing families options to send their children to schools outside of the school district can increase property values in districts with low levels of income or student achievement (Epple and Romano 2003; Nechyba 2000, 2003; Ferreyra 2007), and Brunner, Cho, and Reback (2012) provide empirical evidence to support these predictions. We are not aware of any direct empirical evidence of the impact of charter schools on property values.

4. 

In theory, costs might be reduced due to better resource allocation, more productive use of existing resources, and/or the district operating at a more efficient size after losing student enrollments to charter schools.

5. 

Officials in Buffalo pointed to a study of enrollment patterns in seven charter schools that indicated those schools drew students from 61 different schools and the typical school lost only five students.

6. 

We do not assume the education sector operates with perfect efficiency. If we did, then costs would be defined as the minimum resources required to provide a given level of services and only the unavoidable excess costs would properly be called “costs.”

7. 

Teske et al. (2000) argue that in areas with rising total enrollment, districts avoid any fiscal pain associated with charter school enrollments because they are able to maintain enrollment levels despite declining market share. See also RPP International (2001), which finds that districts with declining enrollment reported that charter schools had a negative impact on their budget, whereas administrators in districts with increasing enrollment trends were more likely to report no fiscal impacts.

8. 

Charter schools in New York are eligible to apply for Title I funding based on their own enrollments.

9. 

Students from low-income families are another group that tends to require additional compensatory services. None of the expenditure categories reported in the data we have access to could be linked explicitly to students from low-income families, however. Thus, our estimates of average variable costs are not adjusted for the socioeconomic composition of charter school students. This limitation of our analysis is likely to result in a small underestimate of the cost savings that could be achieved in Albany, where students from low-income families are overrepresented in charter schools.

10. 

This figure was calculated by multiplying plant operating and maintenance salary expenditures (see appendix table A.1) by 1.283 to account for the costs of benefits and adding the plant operation and maintenance nonpersonnel expenditures and then dividing by fifteen schools. These calculations assume that all facility costs vary directly with the number of schools. That is, there are no fixed overhead costs associated with closed facilities, and thus these might be viewed as upper bound estimates of the additional savings that can be achieved by closing schools.

11. 

District officials in Albany indicated that the recent closure of a middle school in the district created an estimated $750,000 per year in savings, after netting out costs of approximately $150,000 per year associated with maintaining the school as a vacant building. We were also told that the district is in the process of finalizing sale of the building, which will eliminate the costs of “mothballing” the building plus generate fiscal benefits from the sale.

12. 

This figure was calculated by multiplying plant operating and maintenance salary expenditures from appendix table A.2 by 1.331 to account for the costs of benefits and adding the plant operation and maintenance nonpersonnel and then dividing by 58 schools. These calculations assume that all facility costs vary directly with the number of schools, and thus are upper-bound estimates of the additional savings that can be achieved by closing schools.

Acknowledgments

The analysis presented in this brief is based on a longer report (Effect of Charter Schools on School District Finance, Final Report) that the authors submitted to the New York State Education Department and The Education Finance Research Consortium at SUNY Albany on 31 December 2011. The authors would like to thank the New York State Education Research Consortium and the Research Foundation of the State University of New York for funding the research project on which this paper is based, and the New York State Education Department for providing some of the data needed for the project.

REFERENCES

Anderson
,
Amy Berk
.
2004
.
Charter schools in Washington State: A financial drain or gain?
CRPE Working Paper No. 2004_5, University of Washington
.
Arsen
,
David
, and
Yongmei
Ni
.
2012
.
The effects of charter school competition on school district resource allocation.
Educational Administration Quarterly
48
(
1
):
3
38
. doi:10.1177/0013161X11419654
Arsen
,
David
,
David
Plank
, and
Gary
Sykes
.
1999
.
School choice policies in Michigan: The rules matter
.
Lansing, MI
:
Michigan State University
.
Bettinger
,
Eric
.
2005
.
The effect of charter schools on charter students and public schools.
Economics of Education Review
24
(
2
):
133
147
. doi:10.1016/j.econedurev.2004.04.009
Bifulco
,
Robert
, and
Helen F.
Ladd
.
2006
.
The impacts of charter schools on student achievement: Evidence from North Carolina.
Education Finance and Policy
1
(
1
):
50
90
. doi:10.1162/edfp.2006.1.1.50
Booker
,
Kevin
,
Scott
Gilpatric
,
Timothy
Gronberg
, and
Dennis
Jansen
.
2008
.
The effect of charter schools on traditional public school students in Texas: Are children who stay behind left behind?
Journal of Urban Economics
64
(
1
):
123
145
. doi:10.1016/j.jue.2007.10.003
Brunner
,
Eric
,
Sung-Woo
Cho
, and
Randall
Reback
.
2012
.
Mobility, housing markets, and schools: Estimating the effects of interdistrict choice.
Journal of Public Economics
96
(
7
):
604
614
. doi:10.1016/j.jpubeco.2012.04.002
Buddin
,
Richard
.
2012
.
The impact of charter schools on public and private school enrollments
.
Washington, DC
:
Cato Institute
.
Chakrabarti
,
Rajashri
, and
Joydeep
Roy
.
2010
.
Do charter schools crowd out private school enrollment? Evidence from Michigan
.
New York
:
Federal Reserve Bank of New York Staff Report No. 472
.
Epple
,
Dennis
, and
Richard E.
Romano
.
2003
.
Neighborhood schools, choice, and the distribution of educational benefits
. In
The economics of school choice
,
edited by Caroline M. Hoxby
, pp.
227
286
.
Chicago
:
The University of Chicago Press
. doi:10.7208/chicago/9780226355344.003.0008
Ferreyra
,
Maria
.
2007
.
Estimating the effects of private school vouchers in multidistrict economies.
American Economic Review
97
(
3
):
789
817
. doi:10.1257/aer.97.3.789
Goldhaber
,
Dan
,
Kacey
Guin
,
Jeffrey
Henig
,
Frederick
Hess
, and
Janet
Weiss
.
2005
.
How school choice affects students who do not choose
. In
Getting choice right: Ensuring equity and efficiency in education policy
,
edited by Julian Betts and Tom Loveless
, pp.
101
129
.
Washington, DC
:
Brookings Institution
.
Imberman
,
Scott
.
2011
.
The effect of charter schools on achievement and behavior of public school students.
Journal of Public Economics
95
(
7–8
):
850
863
. doi:10.1016/j.jpubeco.2011.02.003
Little
,
David
,
Gregory
Roberts
,
Diane
Ward
,
Alison B.
Bianchi
, and
Mary
Metheny
.
2003
.
Charter schools: investment in innovation or folly?
Latham, NY
:
New York State School Boards Association
.
Ludner
,
Matthew
.
2007
. The impact of charter schools on Catholic schools: A comparison of programs in Arizona and Michigan. Catholic Education: A Journal of Inquiry and Practice 11(1):102–114.
Molnar
,
Alex
.
1996
.
Charter schools: The smiling face of disinvestment.
Educational Leadership
54
(
2
):
9
15
.
National Alliance of Public Charter Schools (NAPCS)
.
2011
.
A growing movement: America's largest charter school communities; Sixth annual edition
.
Washington, DC
:
NAPCS
.
Nechyba
,
Thomas J.
2000
.
Mobility, targeting, and private school vouchers.
American Economic Review
90
(
1
):
130
146
. doi:10.1257/aer.90.1.130
Nechyba
,
Thomas J.
2003
.
Introducing school choice into multidistrict public school systems
. In
The economics of school choice
,
edited by Caroline M. Hoxby
, pp.
145
194
.
Chicago
:
The University of Chicago Press
. doi:10.7208/chicago/9780226355344.003.0006
Ni
,
Yongmei
, and
David
Arsen
.
2010
.
The competitive effect of charter schools on public school districts
. In
The charter school experiment: Expectations, evidence, and implication
,
edited by Chris Lubienski and Peter Weitzel
, pp.
93
120
.
Cambridge, MA
:
Harvard Education Press
.
RPP International
.
2001
.
Challenge and opportunity: The impact of charter schools on school districts
.
Washington, DC
:
U.S. Department of Education, Office of Educational Research and Improvement
.
Sass
,
Tim R.
2006
.
Charter schools and student achievement in Florida.
Education Finance and Policy
1
(
1
):
91
122
. doi:10.1162/edfp.2006.1.1.91
Teske
,
Paul
,
Mark
Schneider
,
Jack
Buckley
, and
Sara
Clark
.
2000
.
Does charter school competition improve traditional public schools?
Civic Report No. 10, Manhattan Institute for Policy Research
.
Toma
,
Eugenia
,
Ron
Zimmer
, and
John T.
Jones
.
2006
.
Beyond achievement: Enrollment consequences of charter schools in Michigan.
Advances in Applied Microeconomics
14
(
2
):
241
255
. doi:10.1016/S0278-0984(06)14009-2

APPENDIX

Table A.1.
Expenditures (in $) by Function in Albany City School District, 2009–10
Step/CostCost per
TotalFixedVariableDriverUnit
Administration      
 General support – salaries 1,829,479 1,829,479   
 General support – non-personnel 3,580,208 3,580,208   
 Plant oper. & maint.-salaries 4,952,447 4,952,447   
 Plant oper. & maint.-non-personnel 5,964,954 5,964,954   
 Superv. & improvement - salaries 9,083,981 2,274,936 6,809,045 Teachers 9,004 
 Superv. & improvement - 1,316,570 1,316,570 Teachers 1,741 
  non-personnel      
Teachers & Instruction (Regular Ed)      
 Kindergarten teacher salaries 2,652,857 2,652,857 K students 3,971 
 G1-G6 teachers salaries 20,173,902 20,173,902 G1-G6 stu. 5,616 
 G7-G12 teachers salaries 19,860,578 19,860,578 G7-G12 stu. 5,412 
 Bilingual/ESL teachers salaries   955,745 955,745 LEP stu. 1,904 
 Non-instructional salaries 3,381,555 3,381,555 All students   424 
 Non-personnel 5,076,845 5,076,845 All students   636 
Special Education      
 Salaries 17,259,348 17,259,348 Stu. w Disabil. 10,005 
 Non-personnel 2,792,079 2,792,079 Stu. w Disabil. 1,619 
Pupil Servicesa  –   
 Salaries 10,756,927 2,150,720 8,606,207 All students 1,348 
 Non-personnel 3,114,031 304,570 2,809,461 All students   381 
Transportation      
 Salaries   166,835   166,835   
 Non-personnel 6,047,031 6,047,031   
Pre-K      
 Salaries 1,311,506 1,311,506   
 Non-personnel 1,065,089 1,065,089   
Other Expendituresb      
 Salaries 3,311,550 1,143,498 2,168,052 G7-G12 stu.   481 
 Non-personnel   853,660   407,949   445,711 G7-G12 stu.   121 
 Capital & debt service 16,027,764 16,027,764   
TOTAL 177,134,243 59,631,447 117,502,796   
Step/CostCost per
TotalFixedVariableDriverUnit
Administration      
 General support – salaries 1,829,479 1,829,479   
 General support – non-personnel 3,580,208 3,580,208   
 Plant oper. & maint.-salaries 4,952,447 4,952,447   
 Plant oper. & maint.-non-personnel 5,964,954 5,964,954   
 Superv. & improvement - salaries 9,083,981 2,274,936 6,809,045 Teachers 9,004 
 Superv. & improvement - 1,316,570 1,316,570 Teachers 1,741 
  non-personnel      
Teachers & Instruction (Regular Ed)      
 Kindergarten teacher salaries 2,652,857 2,652,857 K students 3,971 
 G1-G6 teachers salaries 20,173,902 20,173,902 G1-G6 stu. 5,616 
 G7-G12 teachers salaries 19,860,578 19,860,578 G7-G12 stu. 5,412 
 Bilingual/ESL teachers salaries   955,745 955,745 LEP stu. 1,904 
 Non-instructional salaries 3,381,555 3,381,555 All students   424 
 Non-personnel 5,076,845 5,076,845 All students   636 
Special Education      
 Salaries 17,259,348 17,259,348 Stu. w Disabil. 10,005 
 Non-personnel 2,792,079 2,792,079 Stu. w Disabil. 1,619 
Pupil Servicesa  –   
 Salaries 10,756,927 2,150,720 8,606,207 All students 1,348 
 Non-personnel 3,114,031 304,570 2,809,461 All students   381 
Transportation      
 Salaries   166,835   166,835   
 Non-personnel 6,047,031 6,047,031   
Pre-K      
 Salaries 1,311,506 1,311,506   
 Non-personnel 1,065,089 1,065,089   
Other Expendituresb      
 Salaries 3,311,550 1,143,498 2,168,052 G7-G12 stu.   481 
 Non-personnel   853,660   407,949   445,711 G7-G12 stu.   121 
 Capital & debt service 16,027,764 16,027,764   
TOTAL 177,134,243 59,631,447 117,502,796   

Source: Authors’ computations based on 2009–10 ST-3 and PMF files.

Notes:aIncludes library/instructional technology and food services, which are not typically classified as pupil services.

bIncludes employment preparation, occupational education, special schools, community service, athletics, as well as capital and debt service.

Table A.2.
Expenditures (in $) by Function in Buffalo City School District, 2009–10
Step/CostCost per
TotalFixedVariableDriverUnit
Administration      
 General support – salaries 7,364,408 7,364,408   
 General support – non-personnel 4,206,187 4,206,187   
 Plant oper. & maint.-salaries 16,487,821 16,487,821   
 Plant oper. & maint.-non-personnel 39,755,695 39,755,695   
 Superv. & improvement - salaries 26,896,356 5,207,867 21,688,489 Teachers  6,972 
 Superv. & improvement - 4,406,707 4,406,707 Teachers  1,416 
  non-personnel      
Teachers & Instruction (Regular Ed)      
 Kindergarten teacher salaries 7,059,512 7,059,512 K students  2,894 
 G1-G6 teachers salaries 52,479,365 52,479,365 G1-G6 stu.  3,533 
 G7-G12 teachers salaries 66,209,428 66,209,428 G7-G12 stu.  4,364 
 Bilingual/ESL teachers salaries 7,224,325 7,224,325 LEP stu.  2,349 
 Non-instructional salaries 320,519 320,519 All students   10 
 Non-personnel 23,589,493 23,589,493 All students   723 
Special Education      
 Salaries 62,274,732 62,274,732 Stu. w Disabil. 10,101 
 Non-personnel 5,751,079 5,751,079 Stu. w Disabil.   933 
Pupil Servicesa      
 Salaries 15,767,630 154,339 15,613,291 All students   484 
 Non-personnel 9,023,971 7,558,539 1,465,432 All students   202 
Transportation      
 Salaries 8,453,579 8,453,579   
 Non-personnel 40,948,094 40,948,094   
Pre-K      
 Salaries 7,221,685 7,221,685   
 Non-personnel 1,930,745 1,930,745   
Other Expendituresb      
 Salaries 24,152,178 6,000,644 18,151,534 G7-G12 stu.  1,109 
 Non-personnel 5,464,061 3,321,041 2,143,020 G7-G12 stu.   141 
 Capital & debt service 89,303,537 89,303,537   
TOTAL 680,234,255 308,818,182 371,416,073   
Step/CostCost per
TotalFixedVariableDriverUnit
Administration      
 General support – salaries 7,364,408 7,364,408   
 General support – non-personnel 4,206,187 4,206,187   
 Plant oper. & maint.-salaries 16,487,821 16,487,821   
 Plant oper. & maint.-non-personnel 39,755,695 39,755,695   
 Superv. & improvement - salaries 26,896,356 5,207,867 21,688,489 Teachers  6,972 
 Superv. & improvement - 4,406,707 4,406,707 Teachers  1,416 
  non-personnel      
Teachers & Instruction (Regular Ed)      
 Kindergarten teacher salaries 7,059,512 7,059,512 K students  2,894 
 G1-G6 teachers salaries 52,479,365 52,479,365 G1-G6 stu.  3,533 
 G7-G12 teachers salaries 66,209,428 66,209,428 G7-G12 stu.  4,364 
 Bilingual/ESL teachers salaries 7,224,325 7,224,325 LEP stu.  2,349 
 Non-instructional salaries 320,519 320,519 All students   10 
 Non-personnel 23,589,493 23,589,493 All students   723 
Special Education      
 Salaries 62,274,732 62,274,732 Stu. w Disabil. 10,101 
 Non-personnel 5,751,079 5,751,079 Stu. w Disabil.   933 
Pupil Servicesa      
 Salaries 15,767,630 154,339 15,613,291 All students   484 
 Non-personnel 9,023,971 7,558,539 1,465,432 All students   202 
Transportation      
 Salaries 8,453,579 8,453,579   
 Non-personnel 40,948,094 40,948,094   
Pre-K      
 Salaries 7,221,685 7,221,685   
 Non-personnel 1,930,745 1,930,745   
Other Expendituresb      
 Salaries 24,152,178 6,000,644 18,151,534 G7-G12 stu.  1,109 
 Non-personnel 5,464,061 3,321,041 2,143,020 G7-G12 stu.   141 
 Capital & debt service 89,303,537 89,303,537   
TOTAL 680,234,255 308,818,182 371,416,073   

Source: Authors’ computations based on 2009–10 ST-3 and PMF files.

Notes:aIncludes library/instructional technology and food service, which are not typically classified as pupil services.

bIncludes employment preparation, occupational education, special schools, community service, athletics, as well as capital and debt service.